Wind power interval prediction method based on MODWT-Informer network

The invention discloses a wind power interval prediction method based on an MODWT-Informer network, and belongs to the field of wind power generation output power prediction, and the method comprises the following steps: 1, decomposing original wind power data into different subsequences through employing MODWT, and separating the long-term trend and periodicity of the original wind power data; step 2, adding corresponding original time labels to all decomposed equal-length subsequences, and inputting different subsequences into an independent Informer network for training and prediction; and step 3, carrying out wavelet reconstruction on the values of all the subsequences predicted by the Informer network, synthesizing original wind power data, and completing wind power determinacy prediction. And step 4, using an EnbPI method to expand the reconstructed wind power data into a prediction interval, and completing wind power probabilistic prediction. According to the wind power interval prediction method based